Comment by wrs

Yes, isn’t the whole “alignment” fear basically that if we had smarter than human AGI, we would need smarter than human prompt engineering?